Project Amaze!

Whose speech is the most unusual in terms of parts-of-speech?

In the following table, the higher the score (χ²), the more unusual the proportions of different parts-of-speech are.

Those marked ** have a p-value < 0.01.

In a future beanbag, we’ll break down which parts-of-speech are unusual for certain characters.

Part-of-speech tagging was done with spaCy. The tagging uses Universal POS tags.

Tags PUNCT, SPACE, SYM, and X were ignored.

Only speakers with 100 or more tokens were considered.

The score χ² = Σ (O-E)²/E where O is the observed count and E is the expected count (based on the proportions across all speakers).